SMELL. 231 



Plateau, and Mayor, since the extirpation of the antennae in most cases de- 

 stroyed the sense of smell without rendering the insects otherwise abnormal. 

 Graber's second series of experiments is especially important, since they 

 led him to abandon his former view that the antennae were not olfactory 

 organs. (See also Schenk, 1903 : 573.) 



A scrutiny of Mclndoo's conclusions shows that they are not in entire 

 accord, though this may be due to a desire to avoid overstatement. The 

 statements concerned are as follows: 



"These results indicate that the olfactory organs are located elsewhere. At most 

 it can be claimed only that the antennae may assist in the receiving of odor stimuli" 

 (1914: 297). "It may now be assumed that these pores constitute the olfactory organs 

 in the honey-bee, and perhaps in insects in general" (lb., 341). "Then it will be real- 

 ized that the antennas can no longer be regarded as the seat of the sense of smell 

 in insects" (lb. 345). "In conclusion, it seems that the organs called olfactory pores 

 by the author are the true olfactory apparatus in Hymenoptera and possibly in all 

 insects and that the antennae play no part in receiving odor stimuli" (1914 3 :56). 



The mature bees used in experiments with the antennae cut off gave either 

 greatly delayed reactions or none at all, but when these were burned off, 

 the reaction-time was but slightly increased. However, the latter lived 

 an average of 5% days, while the former in some cases lived an equal period, 

 indicating that both were equally abnormal and that the discrepancy must 

 be sought elsewhere. Forel long ago criticized Graber's results because of 

 the strong odors employed, insisting that natural odors to which the various 

 insects are accustomed should be used, and this objection may hold in some 

 measure for the essential oils. Moreover, it seems possible that reaction- 

 time experiments do not furnish such decisive tests as those made by Forel, 

 Hauser, Mayor, and others. Furthermore, the use of a stop-watch would 

 appear almost imperative, especially for short intervals. 



Mclndoo frequently finds a relation between the number of pores and 

 the reaction-time. "If the reaction-time of each caste of the honey-bee 

 is compared with the total number of olfactory pores, a consistent ratio is 

 obtained. A drone has 2,600 pores and responds in 2.9 seconds; a worker 

 posesses 2,200 pores and responds in 3.4 seconds, and a queen has 1,800 

 pores and responds in 4.9 seconds." This, however, is only a qualitative 

 relation, since the worker with 400 pores less than the drone requires but 

 0.5 second longer, while the queen with 800 pores less takes 2 seconds 

 longer. The absence of a definite quantitative relation is further shown 

 by the fact that removing the wings of workers with 1,510 pores increased 

 the reaction-time 8 times, while coating the 658 pores of the legs increased 

 it but 2.5 times. Finally, although wasps are generally regarded as having 

 a keener sense of smell than bees, the one species studied possessed only 

 1,957 pores. 



On theoretical grounds it is difficult to reconcile the large number of 

 pores and their distribution with the high development of the sense of 

 smell in insects. This is often much more highly developed than vision, 

 and one would expect a highly differentiated organ to correspond and one in 

 immediate connection with the "brain," as is the compound eye. On the 

 contrary, there are hundreds of pores, some of them as far away from the 

 central ganglion as possible and several times as far as others. Moreover, 



